This longitudinal study investigates the causes of relapse following treatment of opiate addiction. Using structured interview schedules, the study team is interviewing 240 subjects at three points in time: upon completion of treatment, after six months, and after twelve months. We will use multivariate statistical analyses (regression, logistic regression and factor analysis) to make causal inferences from the resulting non-experimental data. The analysis seeks to estimate a causal model of relapse that incorporates ideas from the writings of Wikler, Winick, Lindesmith, Marlatt, Brill, Waldorf, Vaillant, and McAuliffe. The results should suggest ways of preventing relapse, and will be especially relevant for the late stages of drug treatment and for aftercare programs. This proposal seeks continued funding for one year to complete twelve-month followup interviews for 75 subjects and to analyze the data from this sample and relevant data from a previous study by the Principal Investigator. The original grant received only 40% of its approved funding in the first-year and subjects have been more difficult to recruit (partly due to red tape) than originally anticipated. With recruitment problems now solved and the followup interviews being completed at an excellent rate, the additional time and money will allow completion of the study as originally planned and analysis of relevant from the previous study.